Zhu et al. Molecular (2019) 12:41 https://doi.org/10.1186/s13039-019-0453-1

RESEARCH C-banding and AgNOR-staining were still effective complementary methods to indentify chromosomal heteromorphisms and some structural abnormalities in prenatal diagnosis Jian Jiang Zhu , Hong Qi* , Li Rong Cai, Xiao Hui Wen, Wen Zeng, Guo Dong Tang, Yao Luo, Ran Meng, Xue Qun Mao and Shao Qin Zhang

Abstract Background: In prenatal diagnosis, CMA has begun to emerge as a favorable alternative to karyotype analysis, but it could not identify balanced translocations, triploidies, inversion and heteromorphisms. Therefore, conventional cytogenetic and specific staining methods still play an important role in the work-up of chromosome anomaly. This study investigated the application of C-banding and AgNOR-staining techniques in prenatal diagnosis of chromosomal heteromorphisms and some structure abnormalities. Results: Among the 2970 samples, the incidence of chromosomal heteromorphisms was 8.79% (261/2970). The most frequent was found to be chromosome Y (2.93%, 87/2970), followed by chromosome 1 (1.65 %, 49/2970), 9 (1.52 %, 45/2970), 22 (0.77 %, 23/2970) and 15 (0.64 %, 19/2970). We compared the incidence of chromosomal heteromorphisms between recurrent spontaneous abortion (RSA) group and control group. The frequency of autosomal hetermorphisms in RSA group was 7.63% higher than that in control group (5.78%), while the frequency of Y chromosomal heteromorphisms was 4.76% lower than that in control group (5.71%). Here we summarized 4 representative cases, inv (1) (p12q24), psu dic (4;17) (p16.3;p13.3), r(X)(p11; q21) and an isodicentric bisatellited chromosome to illustrate the application of C-banding or AgNOR-staining, CMA or NGS was performed to detect CNVs if necessary. Conclusions: This study indicated that C-banding and AgNOR-staining were still effective complementary methods to identify chromosomal heteromorphisms and marker chromosomes or some structural rearrangements involving the centromere or acrocentric chromosomes. Our results suggested that there was no evidence for an association between chromosomal heteromorphisms and infertility or recurrent spontaneous abortions. Undoubtedly, sometimes we needed to combine the results of CMA or CNV-seq to comprehensively reflect the structure and aberration of chromosome segments. Thus, accurate karyotype reports and genetic counseling could be provided. Keywords: Prenatal diagnosis, C-banding, AgNOR-staining, Chromosomal heteromorphisms, Chromosomal structural abnormality, Recurrent spontaneous abortion

* Correspondence: [email protected] Prenatal Diagnosis Center, Beijing Haidian Maternal and Child Health Hospital, Beijing 100080, People’s Republic of China

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 2 of 11

Introduction based on the indications of high-risk including advanced Currently, chromosomal microarray analysis (CMA) and maternal age, positive aneuploidy screening results, or copy number variation sequencing (CNV-Seq) are widely ultrasound structural anomalies in the Beijing Haidian used in prenatal diagnosis due to the capability of identi- Maternal and Child Health Hospital. The pregnant fying microdeletion and microduplication syndromes as women’s obstetrical history was detail recorded, 2970 suc- well as de novo pathogenic CNVs which may be missed by cessful diagnostic samples were divided into two groups: conventional karyotyping [1]. However, their drawbacks the RSA group (history of spontaneous abortions≥2, are also obvious including the incompetence of balanced Group A) and the control group (history of spontaneous translocations, triploidies, inversion and heteromorphisms abortions≤1, Group B). Samples with chromosomal het- identification. So far, most of the clinical experts do not eromorphisms, structural abnormalities involving the support the replacement of conventional karyotyping by centromere regions or acrocentric chromosomes, as well CMA as a first-tier clinical test or in pregnancies at low as mark chromosomes were verified by C-banding and risk for chromosome anomalies [2, 3] mainly due to the AgNOR-staining techniques at first. Then CMA or NGS detection of variant of unknown significance (VOUS) and was also used to exclude chromosome CNVs if necessary. economic reasons. Therefore, the karyotype analysis All clinical tests were based on written informed consent technology is still essential for chromosomal disorder of pregnant women. diagnosis. After different treatments such as denaturation and/or Cell culture and karyotype analysis enzymatic digestion, chromosomes show light and dark Fetal cells obtained from villus, amniotic fluid or cord bands under light microscopy in different regions. blood were cultured with double-line using standard Chromosome banding techniques are generally divided methodologies. Chromosomes in metaphases were into two types: (1) bands are distributed along the entire prepared from the cultured cell or lymphocytes using chromosome, e.g. the most frequently used Giemsa-tryp- Giemsa-trypsin banding techniques. At least twenty sin banding (G-banding), Q banding and R banding; (2) metaphases chromosomes were counted and three meta- bands are located in specific chromosomal regions, phases chromosomes were analyzed in each line by two including C-banding (constitutive heterochromatin), Ag- independent laboratory technicians to avoid uncertainty NOR staining (nucleolar organizer region) and T-banding and variable results. (telomere). Each chromosome has a unique band pattern Chromosomal heteromorphisms were described follow- that can be reliably identified according to the corre- ing the criteria in the International System for Chromo- sponding banding technique. However, a dark-stained some Nomenclature [10]. 1/9/16qh+ is defined as being at band by a particular treatment may appear light-stained in least twice the length of the corresponding region on the another. G-banding is the most frequently used method in homologous chromosome, qh- is defined as being half the clinical laboratories because of its stability and cost- length of the normal corresponding region. Polymor- effectiveness. phisms of the Y chromosome were evaluated as such that C-banding and AgNOR-staining as the main banding Yqh+ (>size of chromosome 18), and Yqh- (

Table 1 Summary the number and frequency of all observed Chromosomal heteromorphisms Heteromorphisms Chromosome number types 1 9 16 19 13 14 15 21 22 Y Total (n; %) qh+ 44 15 10 1 - - - - - 32 102 (3.43%) qh- 5 2 0 0 - - - - - 49 56 (1.89%) ps+ - - - - 1 - 7 9 14 - 31 (1.04%) pss - - - - - 1 - 1 1 - 3 (0.1%) pstk+ - - - - 3 5 1 4 5 - 18 (0.61%) cenh+/- - - - - 0/1 1/0 11/0 0/1 3/0 - 17 (0.57%) inv- 28------6 34(1.14%) Total (n; %) 49 (1.65%) 45 (1.52%) 10 (0.34%) 1 (0.03%) 5 (0.17%) 7 (0.24%) 19 (0.64%) 15 (0.51%) 23 (0.77%) 87 (2.93%) 261 (8.79%)

Ag-NOR staining CMA and CNV-seq The slide was covered with 4 layers of clean lens paper Affymetrix CytoScan 750K array platforms were used to (the size was slightly smaller than the slide) first. Then, detect copy number variants. Genomic DNA extraction it was moved to a plate floating in a 65°C water bath. was performed using Genomic DNA Extraction kit 5ml of freshly prepared 10% AgNO3 solution (containing (QIAamp DNA Blood Mini Kit, QIAGEN GmBH, 0.1% formic acid) was dropped onto the lens paper sev- Hilden, Germany) according to the in-house protocols. eral times by a straw and the slide remained in the water The standard experimental procedure incorporated the bath for about 15min until the lens paper turned black following steps: digestion, ligation, polymerase chain re- or dark brown. After gently removing the lens paper action (PCR), PCR purification, fragmentation, labeling, with tweezers, the slide was washed with tap water and hybridization, washing, staining and scanning. Data was samples were finally stained with Giemsa for 1 minute in analyzed using the Chromosome analysis software room temperature. The most important thing in the (Chromosome Analysis Suite version 2.1) (Affymetrix; whole process of the experiment was to avoid the light. Thermo Fisher Scientific, Inc.). Copy number variation

Fig. 1 Partial karyotypes of chromosomal heteromorphisms by G-banding and C/AgNOR-staining Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 4 of 11

Table 2 Number and frequency of autosomal heteromorphisms in different groups Group Chromosome heteromorphisms No. of cases 1/9/16/19qh± D/G variation inv(9) Total of heteromorphisms (n; %) Chi-square test Group Aa 131 4 (3.05%) 4 (3.05%) 2 (1.53%) 10 (7.63%) Pearson χ2=0.783 p>0.05 Group Bb 2839 73 (2.57%) 65 (2.29%) 26 (0.92%) 164 (5.78%) Total 2970 77 (2.59%) 69 (2.32%) 28 (0.94%) 174 (5.86%) *Significant at p < 0.05. athe recurrent spontaneous abortion group bthe control group sequencing (CNV-seq) was performed as previously chromosomal heteromorphisms, followed by Yqh- 3.26 described [11, 12]. %, Yqh+ 2.09 %, and inv(Y) 0.39% (Table 3). The partial The conventional genomic and phenotype public data- karyotypes of chromosomal heteromorphisms by G- bases such as UCSC Genome Browser (http://genome. banding and C/AgNOR-staining are presented in Fig. 1. ucsc.edu/cgi-bin/hgGateway), ClinGene (http://www. Parental studies had been performed for most prenatally clinicalgenome.org/), OMIM (http://omim.org), DE- reported chromosome heteromorphisms, particularly for CIPHER (https://decipher.sanger.ac.uk), DGV (http:// pericentric inversion, confirming that these variations dgv.tcag.ca/dgv/app/home) and PubMed (http://www. were stably inherited in family. ncbi.nlm.nih.gov/pubmed) were used for retrieval and The group A had a higher frequency (7.63%) of auto- interpretation. somal hetermorphisms compared with the group B (5.78%)(Table 2), but a lower frequency (4.76%) of Y Statistical analysis chromosomal heteromorphisms compared with the group Data were analyzed using SPSS Statistics (version 22.0). B (5.71%) (Table 3). While there were no statistically Differences between the RSA and the control groups significant difference between two groups (P >0.05). within the cohort were tested with Chi-squared test statistics. The significance level was set at p < 0.05. Chromosome abnormal C-banding and AgNOR-staining were important banding Results methods to characterize marker chromosomes or other Among the 2972 prenatal diagnosis cases (including structural rearrangements involving the centromere or 2735 amniotic fluid, 68 villus and 169 cord blood), 2 acrocentric chromosomes. Here we summarized 4 repre- cases of amniotic fluid failed to culture due to less sentative cases, inv (1) (p12q24), psu dic (4;17)(p16.3; cloning, and the success rate of cultivation was 99.93% p13.3), r(X)(p11;q21) and an isodicentric bisatellited (the proportion of male and female was 1533:1437). A chromosome to illustrate the application of C-banding total of 131women were investigated in the group A, or AgNOR-staining. CMA or NGS was performed to and the group B consisted of 2839. detect CNVs. Among them , 2 CNVs with VOUS were detected in case#1, pathogenic CNVs were detected in Chromosome heteromorphisms case #2 and case #3, respectively (Table 4). The distribution among different chromosomes hetero- morphisms was presented in Table 1 with total fre- Discussion quency of 8.79% (261/2970). The most frequent was Chromosomal variation were mainly refers to the found to be chromosome Y (2.93%, 87/2970), followed variations on heterochromatic segments, satellites and by chromosome 1 (1.65 %, 49/2970), 9(1.52 %, 45/2970), satellite stalks in the population, but euchromatic 22 (0.77 %, 23/2970) and chromosome 15 (0.64 %, 19/ heteromorphisms with C-banding and Ag-NOR negative 2970). In 1533 male fetuses, 5.68 % (87/1533) had Y could also be included [13–15]. Like most other

Table 3 Number and frequency of Y Chromosomal heteromorphisms in different groups Group Chromosome heteromorphisms No. of male cases Yqh+ Yqh- inv (Y) Total of heteromorphisms (n; %) Chi-square test Group Aa 63 0 2 (3.17%) 1 (1.59%) 3 (4.76%) Continuity Correction χ2=0.002 p > 0.05 Group Bb 1470 32 (2.18%) 47 (3.2%) 5 (0.34%) 84 (5.71%) Total 1533 32 (2.09%) 49 (3.2%) 6 (0.39%) 87 (5.68%) *Significant at p < 0.05 athe recurrent spontaneous abortion group bthe control group Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 5 of 11

Table 4 C-banding or AgNOR-staining applied to the auxiliary diagnosis of 4 cases with chromosome abnormal Case No. Brief clinical information Chromosome karyotype SNP-array/CNV-seq Pregnancy outcome Case#1 42-year-old, G5P1,amniocentesis at 19 weeks’ F:46,XN,inv(1)(p12q24)mat F::arr [hg19] 2q13(110,498,141_ continued gestation because of advanced age P: 46,XX,inv(1)(p12q24) 110,980,295)x4, pregnancy H:46,XY 12p12.1 (23,797,551_24,076, 457)x1 P: arr [hg19] 2q13(110,498,141_ 110,980,295)x4 H: arr(1-22)x2,(XN)x1 Case#2 25-year-old, G5P1,amniocentesis at 18 weeks’ F:45,XY, F:arr [hg19] 2q12.3q13(107,586, induced gestation because of a positive serological screening der(4)dup(4)(p15.1p16) 661_110,980,295)x3, abortion result (a high risk of Down syndrome 1:176) psu dic(4;17)(p16.3;p13.3),- 4p16.3p15.1 (68,345_32,437, 17 069)x3 P:46,XX H:46,XY Case#3 35-year-old, G4P1,puncture of umbilical vein at 26 weeks’ F:45,X [31]/46,X,r(X)(p11; F: induced gestation because of advanced age and ultrasonic anomalies q21) [29] seq[GRCh37]del(X)(p22.33p11.3), abortion (coarctation of the aorta? long bone dysplasias). P:46,XX del(X)(q21.31q28) H:46,XY ChrX:g.60001_43660000del, 0710001_155260000del Case#4 37-year-old, G2P1,amniocentesis at 19 weeks’ gestation F:47,XN,+mar [37]/46, F:arr(1-22)x2,(XN)x1 continued because of advanced age XN[63] pregnancy P:46,XX H:46,XY Abbreviations: F Fetus, P Pregnant woman, H Husband of pregnant woman literatures, here we mainly focused on the C-banding or the literature was consistent with ours [8, 17], some did AgNOR-staining positive chromosomal heteromorphisms. not mention the details of the judgment criteria [9, 18]. As anticipated, while most of chromosmal hetero- This could be one of the reasons for conflicting reports of morphisms were successfully detected according to heteromorphisms impact human reproduction. Shivanand our criteria, we still occasionally encounter disputes [16] considered that the length of short arms of chromo- between experienced technicians occasionally. Without a some 16 was not appreciably altered by compaction, reference standard other than between homologues, it is Moreover, it is intermediate in size compared to the qh re- difficult to determine whether two similar-sized homo- gions of 1, 9,and 16. Also, it is easily identified in a C- logues are slightly smaller or larger than the normal ones banded cell, therefore provides a useful reference standard [16]. In addition, we found that the diagnostic criteria in to set criteria. Frequencies of heteromorphisms in various different literatures were not uniform, although most of populations showed differences due to ethnic origins, age

Fig. 2 Prenatal diagnosis of chromosome 19 heteromorphism by G-banding (a) and C-banding (b). Chromosome 19 is labelled, with an arrow indicating the large heterochromatic region Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 6 of 11

and geographical distribution. Like other studies of Asian emphasize the ultimate importance of studying heterogen- populations had shown greater variation in the size of the eity in biology, including heteromorphisms and euchro- Y than in the white population [19]. In our study, the fre- matic variants [20]. Of equal importance, CNV-seq and quency of Y chromosome heteromorphisms was the high- CMA will complement the karyotyping method by allow- est, which was apparently higher than other literature [6] ing the detetion of small CNV, with more potentially and Yqh-was slightly more common than Yqh+. Perhaps pathogenic genetic candidates detected among the CNV the populations selected introduced biases and the criteria regions, refined genetic causes of conditions like spontan- used in current studies were subjective, making it difficult eous miscarriage and other medical syndromes could be to directly compare frequencies. Our data showed that investigated [12, 21]. each frequency of chromosomal heteromorphisms does In recent years, more and more studies had focused not appear to be significantly different in group A and on the chromosomal heteromorphisms [22] which can group B, confirmed that variation in these C-banding or be verified by the application of C-banding and AgNOR- AgNOR-staining positive regions has no clinical signifi- staining techniques. Several other centromeric variants, cance. It was almost certain that the common heteromor- verified by C-banding in the course of a prenatal diagnosis phisms were stably inherited in the family by study of the [23–28], have been described for chromosomes 5, 6, 12, parental chromosomes, which would strengthen the argu- 18, 19 or 20. Our data presented a rare heteromorphism ment against the associated between heteromorphisms on chromosome 19 and it was different from the four and infertility. However, further research is required to de- different classes that Crossen noted [29]. A distinct dark- lineate the mechanisms impacting human reproduction, staining band at the long arm of chromosome 19 near the focusing on patients with unexplained infertility, poor em- centromere region was revealed by G-banding (Fig. 2a) bryonic development, and spontaneous abortions. Current and it was also dark-stained by C-banding as the same as cytogenetics has mainly focused its efforts on the identifi- centromere (Fig. 2b). So it was most likely constitutive cation of clonal chromosomal aberrations (CCAs). How- heterochromatin which was considered without patho- ever, many investigators have demonstrated that “non logical significance. As the fetus has no abnormal ultra- clonal chromosome aberrations,” or NCCAs are not sound findings except for increased nuchal translucency “noise” but rather a highly significant feature of the gen- (3.7mm) at 13 weeks’ gestation, the parents rejected fur- ome system, and the significance of NCCAs will ther chromosome verification and CNVs testing. Our

Fig. 3 Prenatal diagnosis of 46,XX,inv (1)(p12q42) in Case #1. Chromosome 1 is labeled, with an arrow indicating the abnormal region. (a) G-banding (Left) and C-banding karyotypes (Right). (b) SNP-array analysis revealed 2q13 double duplication of and 12p12.1deletion of Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 7 of 11

karyotype report was 46, XN, 19qh+. No abnormalities (1)(p12q42) (Case #1, Fig 3), the qh regions of inv (1) were found in the 6-month follow-up after birth. Indeed, appeared in the distal long arm of the 1 chromosome the discovery of rare chromosomal polymorphisms raised rather than in the middle were especially visible by C- the question of whether or not differences in size and banding (Fig. 3a), two VOUS were revealed by Array- banding pattern observed between homologue could SNP (Fig. 3b). After genetic counseling, the pregnant account for the existence of a normal variant. Therefore, woman chose to continue pregnancy with some misgiv- for rare chromosomal heteromorphisms, family verifica- ings. Fortunately, the child had no abnormalities since tion and original analysis should be performed to better birth at term. We will continue to monitor the child's assess the genetic effects of the heteromorphisms. In follow-up to check whether these VOUS really didn’t addition, whether the heteromorphisms will change (such have clinical significance. as changes in the length or size of heteromorphisms re- As shown in the diagnosis of case #2, C-banding can gions) during the genetic processes remains to be further also be used to identify pseudodicentric chromosome studied. and to determine whether a fragment of unknown origin In our prenatal diagnosis work, C-banding was per- is heterochromatin. G-banding showed an apparent formed to verify the pericentric inversion and structure unknown dark-stained band (Fig. 4a, indicated by arrow) rearrangement involving the centromere regions. It was inserted at the breakpoint of the derived chromosome convenient and easy to make diagnosis by observing the formed by reciprocal translocation of chromosomes 4 position and morphological changes of centromeres with and 17. The C-banding demonstrated that the derivative C-banding. Our data showed that several cases of inv (9) chromosome was dicentric with only one primary cannot be clearly diagnosed by G-banding due to the constriction (presumably the active centromere), short and poor karyotype, but it can be easily confirmed whereas the other regions were light stained (Fig. 4b). So by C-banding (Fig. 1). Similarly, in the diagnosis of inv we confirmed that the inserted fragment was not

Fig. 4 Prenatal diagnosis of 45,XY,der (4) dup (4)(p15.1p16) psu dic (4;17)(p16.3;p13.3),-17 in Case #2. Abnormal chromosome is indicated by arrows. (a) G-banding karyotype. (b) C-banding karyotype. (c) The SNP-array analysis revealed 2q12.3q13 and 4p16.3p15.1 duplications Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 8 of 11

constitutive heterochromatin. Therefore, we conducted an addtional cell line with 46 chromosomes due to the the CMA test and finally revealed that it was a 32.3 Mb presence of an extra ring chromosome X [30]. Ring X duplication of 4p16.3p15.1, at the same time, a 3.3Mb chromosome, with different degree of mosaicism, show- duplication of 2q was also detected (Fig. 4c). The ing common clinical characteristics of Turner syndrome pregnancy was terminated at 23 weeks of gestation due had been reported, and karyotype-phenotype correlation to the chromosome abnormality. related manifestation was dependent on the degree of C-banding and AgNOR-staining were also routinely genetic material lost in ring (X) formation as well as used to characterize marker chromosomes. In case #3, mosaicism [31]. Undoubtedly, for mosaic Tuner with Karyotyping revealed mosaic of the mark chromosome mixed cell lines, FISH tests for Xcen and Ycen should be with two types of cell lines 45,X and 46,X,+mar (Fig. 5a). performed since ring X is more commonly seen in The C-banding showed that the mark chromosome had mosaic Turner and present of Y material will be clinic- a clear dark-staining centromere and the rest of the ally significant and requires intervention. In addition, the chromosome regions were light stained (Fig. 5b). There- interphase FISH of uncultured cells can avoid the influ- fore, we hypothesized that the mark could not be the ence of cell culture and thus more accurately identify heterochromatic region of the Y chromosome but likely the degree of mosaicism, the metaphase FISH can clearly to be a circle X with a large fragment deletion, and this locate the recombinant chromosome. CNV-seq or CMA was later confirmed by the CNV-seq analysis (Fig. 5c). analysis can reveal exact break points on both arms of X About 16% of the individuals with Tuner syndrome have chromosomes. In contrast to case #3, the supernumerary

Fig. 5 Prenatal diagnosis of 45,X [31]/46,X,r(X)(p11;q21) [29] in Case #3. Abnormal chromosome is indicated by arrows. (a) G-banding Karyotype. (b) C-banding Karyotype . (c) CNV-seq analysis revealed large fragment deletions in chromosome X deletion. The gray baseline beside the chromosome represent the copy number of chromosomes was normal, green portions of the baseline represented the regions in chromosome were deletions Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 9 of 11

marker chromosomes (SMC) of case #4, which we highly of children’s development, the initial diminished per- speculated that it was a bisatellited metacentric micro- ception of child competency and later dissatisfaction chromosome (Fig. 6a). The further identified by AgNOR- with genomic testing indicate the need to assist par- banding showed prominent satellites on both sides of the ents in coping with ambiguous results [36]. It is sug- marker (Fig. 6b). About 70% of SMC are derived from ac- gested to not change the current policy of microarray rocentric chromosomes and those markers derived from application in prenatal diagnosis until more data on can be identified by FISH [32]. the clinical significance of copy number changes are CMA has advantages over conventional cytogenetic, available [2]. So, without professional genetic counsel- including the ability to precisely characterize CNVs ing qualifications and informed consent of subjects, it associated with abnormal karyotypes. Moreover, a sig- is not appropriate to blindly expand the scope of de- nificant proportion of cases studied by array detected tection without clinical indications. We recommend clinically significant CNVs even in samples with ap- that CMA or CNV-seq used for fetal chromosomal parently normal karyotypes [33, 34], but the expensive structural abnormalities (especially de novo), ultra- inspection costs and VOUS still require us to con- sound abnormalities, and the SMCs with C-banding sider its cost-effectiveness. These scenarios present a and Ag-NOR negative. We also strongly recom- challenge for prenatal diagnosis, and genetic counsel- mended de novo SMC for CNVs examination (case ing prior to prenatal CMA greatly facilitates delivery #4) because that a prospective study showed that 69% of complex results [35]. Although VOUS results have of de novo SMC contained euchromatin material, limited impact on parental well-being and perception 95.4% of which for non-acrocentric markers [37].

Fig. 6 Prenatal diagnosis of 47,XN,+mar [37]/46,XN [63] in Case#4. There is no abnormal detected by SNP-array (Data not shown). The marker chromosome is indicated by arrows. (a) G-banding Karyotype. (b) AgNOR-staining Karyotype Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 10 of 11

Conclusion Authors’ contributions Each technology has its limitations, we didn’t demon- JJZ conceived of the study and performed the statistical analysis, and helped to draft the manuscript. HQ participated in the design of the study, and strate that C-banding and/or AgNOR-staining was ab- participated in its design and coordination. LRC, XHW, WZ, GDT and YL solutely necessary above other available technologies carried out the cytogenetics and molecular genetic studies. RM, XQM, SQZ such as FISH or CMA. But base on our work, it is carried out genetic counseling and interventional prenatal diagnosis for pregnant women. All authors read and approved the final manuscript. suggested that C-banding and AgNOR-staining, which were more convenient, fast and economical, could still Funding be effective complementary methods to study hetero- Not applicable morphic variations, and to characterize marker chro- mosomes or other structural rearrangements involving Availability of data and materials Data supporting the results reported in the published article can be found in centromere regions or acrocentric chromosomes, es- the tables and figures. pecially when it was difficulties to make a definite diagnosis due to the short and/or poor G-banding Ethics approval and consent to participate The sample-related data analyzed for this manuscript was entirely retrospective karyotype. On the other hand, purely using G-banding with no patient or patient-related identifiers included in the analysis. No waivers without combining C-banding and AgNOR can hardly or institutional approval was required. Each patient received written informed tell the difference between cenh+ and ps+, as well as consent for participation. pss and pstk+, and thereby causes the unreliable Consent for publication karyotype results. Informed written consent was obtained from participants for publication for Unsurprisingly, further studies are needed to delineate images and other clinical information relating to these cases to be reported whether heteromorphisms impact human reproduction. for academic purpose. Absolutely, there is an urgent need to establish normal- Competing interests ized diagnostic criteria for chromosomal heteromor- The authors declare that they have no competing interests. phisms, and this is essential if any degree of comparison is to be made between these and other clinical studies. Received: 3 April 2019 Accepted: 23 August 2019 Contrary to some previous studies, we found no evi- dence for an association between chromosomal hetero- References morphisms and infertility or RSA. Our study agreed 1. Wapner RJ, Martin CL, Levy B, Ballif BC, Eng CM, Zachary JM, et al. that familial cytogenetic studies do not appear to be Chromosomal microarray versus karyotyping for prenatal diagnosis [J]. N Engl J Med. 2012;367(23):2175–84. routinely necessary when a common chromosomal het- 2. Miny P, Wenzel F, Tercanli S, Filges I. Chromosomal Microarrays in Prenatal eromorphisms is diagnosed prenatally [38]. Perhaps Diagnosis: Time for a Change of Policy?[J]. Microarrays (Basel). 2013;2(4):304–17. most critically, followup studies are distinctly lacking. 3. Novelli A, Grati FR, Ballarati L, Bernardini L, Bizzoco D, Camurri L, et al. Microarray application in prenatal diagnosis: a position statement from the Studies of the progeny (or products of conception) of cytogenetics working group of the Italian Society of Human carriers of heteromorphic variants are sorely needed to (SIGU), November 2011[J]. Ultrasound Obstet Gynecol. 2012;39(4):384–8. better delineate the heritability and consequences of 4. Akbas H, Isi H, Oral D, Turkyilmaz A, Kalkanli-Tas S, Simsek S, et al. Chromosome heteromorphisms are more frequent in couples with these variants [22]. recurrent abortions [J]. Genet Mol Res. 2012;11(4):3847–51. In summary, in prenatal diagnosis, it is necessary make 5. Wang Y, Li G, Zuo MZ, Fang JH, Li HR, Quan DD, et al. Y chromosome a diagnosis in a limited time according to the limited polymorphisms may contribute to an increased risk of male-induced unexplained recurrent miscarriage [J]. Biosci Rep. 2017;37:2. clinical manifestations of the fetus, so reasonable 6. Dai R, Pan Y, Fu Y, Liu Q, Han W, Liu R. Role of male genetic factors in detection methods should be selected and avoided the recurrent pregnancy loss in Northeast China [J]. Eur J Obstet Gynecol unnecessary expansion of detection scope which gener- Reprod Biol. 2018;224:6–11. 7. Cheng R, Ma Y, Nie Y, Qiao X, Yang Z, Zeng R, et al. Chromosomal ated additional and more complex data to amplify and polymorphisms are associated with female infertility and adverse exacerbate some pre-existing ethical problems [39]. reproductive outcomes after infertility treatment: a 7-year retrospective study [J]. Reprod Biomed Online. 2017;35(1):72–80. 8. Hong Y, Zhou YW, Tao J, Wang SX, Zhao XM. Do polymorphic variants of Abbreviations chromosomes affect the outcome of in vitro fertilization and embryo CCAs: Clonal Chromosomal Aberrations; CMA: Chromosomal Microarry transfer treatment?[J]. Hum Reprod. 2011;26(4):933–40. Analysis; CNV-Seq: Copy Number Variation Sequencing; FISH: Fluorescence In 9. Dong Y, Jiang YT, Du RC, Zhang HG, Li LL, Liu RZ. Impact of chromosomal Situ Hybridization; NCCAs: Non Clonal Chromosome Aberrations; heteromorphisms on reproductive failure and analysis of 38 heteromorphic NOR: Nucleolar Organizer Region; RSA: Recurrent Spontaneous Abortion; pedigrees in Northeast China [J]. J Assist Reprod Genet. 2013;30(2):275–81. SMC: Supernumerary Marker Chromosomes; VOUS: Variant Of Unknown 10. International Standing Committee on Human Cytogenetic Nomenclature, Significance Shaffer LG, Mc Gowan-Jordan J, Schmid M. ISCN 2013 : an international system for human cytogenetic nomenclature (2013)[M]. Basel: Karger; 2013. vi, 140 p. , 141 folded sheet Acknowledgements 11. Liang D, Lv W, Wang H, Xu L, Liu J, Li H, et al. Non-invasive prenatal testing The authors would like to acknowledge and thank all the families for their of fetal whole chromosome aneuploidy by massively parallel sequencing [J]. cooperation and staff who participated in this study. Thank Annoroad Gene Prenat Diagn. 2013;33(5):409–15. Technology and Be reative lab (Beijing) for their expert laboratory work and 12. Qi H, Xuan ZL, Du Y, Cai LR, Zhang H, Wen XH, et al. High resolution global analyses. chromosomal aberrations from spontaneous miscarriages revealed by low Zhu et al. Molecular Cytogenetics (2019) 12:41 Page 11 of 11

coverage [J]. Eur J Obstet Gynecol Reprod Biol. 37. Marle N, Martinet D, Aboura A, Joly-Helas G, Andrieux J, Flori E, et al. 2018;224:21–8. Molecular characterization of 39 de novo sSMC: contribution to prognosis and 13. Jalal SM, Schneider NR, Kukolich MK, Wilson GN. Euchromatic 16p+ genetic counselling, a prospective study [J]. Clin Genet. 2014;85(3):233–44. heteromorphism: first report in North America [J]. Am J Med Genet. 38. Hsu LY, Benn PA, Tannenbaum HL, Perlis TE, Carlson AD. Chromosomal 1990;37(4):548–50. polymorphisms of 1, 9, 16, and Y in 4 major ethnic groups: a large prenatal 14. Webb GC, Krumins EJ, Eichenbaum SZ, Voullaire LE, Earle E, Choo KH. study [J]. Am J Med Genet. 1987;26(1):95–101. Non C-banding variants in some normal families might be 39. Clarke AJ, Wallgren-Pettersson C. Ethics in genetic counselling [J]. J homogeneously staining regions [J]. Hum Genet. 1989;82(1):59–62. Community Genet. 2018. 15. Song XH, Hsu HK, Su MT, Chang TS, Su PY, Chen M, et al. Euchromatic variants of 8q21.2 in twins [J]. Taiwan J Obstet Gynecol. 2017;56(2):227–9. ’ 16. Patil SR, Lubs HA. Classification of qh regions in human chromosomes 1, 9, Publisher sNote Springer Nature remains neutral with regard to jurisdictional claims in and 16 by C-banding [J]. Hum Genet. 1977;38(1):35–8. published maps and institutional affiliations. 17. Sun L, Chen ZH, Yang L, Yi CX, Liu J, Ou CQ. Chromosomal polymorphisms are independently associated with multinucleated embryo formation [J]. J Assist Reprod Genet. 2018;35(1):149–56. 18. Inan C, Sayin NC, Dolgun ZN, Gurkan H, Erzincan SG, Uzun I, et al. Prenatal diagnosis of chromosomal polymorphisms: most commonly observed polymorphism on Chromosome 9 have associations with low PAPP-A values()[J]. J Matern Fetal Neonatal Med. 2017:1–8. 19. Hou JW, Wang TR. Study of human Y chromosome polymorphism in Taiwan [J]. Acta Paediatr Taiwan. 1999;40(5):302–4. 20. Heng HH, Regan SM, Liu G, Ye CJ. Why it is crucial to analyze non clonal chromosome aberrations or NCCAs?[J]. Mol Cytogenet. 2016;9:15. 21. Bagheri H, Mercier E, Qiao Y, Stephenson MD, Rajcan-Separovic E. Genomic characteristics of miscarriage copy number variants [J]. Mol Hum Reprod. 2015;21(8):655–61. 22. Tempest HG, Simpson JL. Why are we still talking about chromosomal heteromorphisms?[J]. Reprod Biomed Online. 2017;35(1):1–2. 23. Fineman RM, Issa B, Weinblatt V. Prenatal diagnosis of a large heteromorphic region in a chromosome 5: implications for genetic counseling [J]. Am J Med Genet. 1989;32(4):498–9. 24. Jabs EW, Carpenter N. Molecular cytogenetic evidence for amplification of chromosome-specific alphoid sequences at enlarged C-bands on chromosome 6[J]. Am J Hum Genet. 1988;43(1):69–74. 25. Chantot-BastaraudS,SiffroiJP,BerkaneN,HeimN,HerveF,UzanS,et al. Prenatal diagnosis of a large centromeric heteromorphism of chromosome 12: implications for genetic counseling [J]. Fetal Diagn Ther. 2003;18(2):111–3. 26. Pittalis MC, Santarini L, Bovicelli L. Prenatal diagnosis of a heterochromatic 18p+ heteromorphism [J]. Prenat Diagn. 1994;14(1):72–3. 27. Friedrich U. Centromere heteromorphism in chromosome 19[J]. Clin Genet. 1985;28(4):358–9. 28. Petersen MB. Rare chromosome 20 variants encountered during prenatal diagnosis [J]. Prenat Diagn. 1986;6(5):363–7. 29. Crossen PE. Variation in the centromeric banding of chromosome 19[J]. Clin Genet. 1975;8(3):218–22. 30. Jacobs P, Dalton P, James R, Mosse K, Power M, Robinson D, et al. Turner syndrome: a cytogenetic and molecular study [J]. Ann Hum Genet. 1997;61(Pt 6:471–83. 31. Chauhan P, Jaiswal SK, Lakhotia AR, Rai AK. Molecular cytogenetic characterization of two Turner syndrome patients with mosaic ring X chromosome [J]. J Assist Reprod Genet. 2016;33(9):1161–8. 32. Dutta UR, Vempally S, Ranganath P, Dalal A. A novel combined 15q11.2 duplication and a bisatellited supernumerary marker derived from chromosome 22: molecular characterization of the marker [J]. Gene. 2014;539(1):162–7. 33. Shaffer LG, Dabell MP, Fisher AJ, Coppinger J, Bandholz AM, Ellison JW, et al. Experience with microarray-based comparative genomic hybridization for prenatal diagnosis in over 5000 pregnancies [J]. Prenat Diagn. 2012;32(10):976–85. 34. Hay SB, Sahoo T, Travis MK, Hovanes K, Dzidic N, Doherty C, et al. ACOG and SMFM guidelines for prenatal diagnosis: Is karyotyping really sufficient?[J]. Prenat Diagn. 2018;38(3):184–9. 35. Levy B, Wapner R. Prenatal diagnosis by chromosomal microarray analysis [J]. Fertil Steril. 2018;109(2):201–12. 36. Desai P, Haber H, Bulafka J, Russell A, Clifton R, Zachary J, et al. Impacts of variants of uncertain significance on parental perceptions of children after prenatal chromosome microarray testing [J]. Prenat Diagn. 2018; 38(10):740–7.